Connecting Big Data for biodiversity monitoring

16 May 2019 09:00
Dr Alex Bush in the field.

New Lancaster researcher combines a novel DNA sequencing technique with aerial images of the earth to monitor and predict how biodiversity will respond to environmental change.

When ecologist Dr Alex Bush first heard of a new, faster, cheaper way to potentially identify all species in a scientific sample - it seemed like an answer to his prayers.

“I went to a conference in 2013 and heard my future supervisor talking about metabarcoding, and it sounded like magic,” said Alex, who was doing his PhD in Australia just as the technique was being developed. Alex had been modelling how species, particularly insects, adapt to climate change, but was reliant on historic data and deeply frustrated by the lack of up to date information.

“I was building better and better models, but still using the same old records and constantly limited by what information I could practically collect ,” said Alex who has just joined Lancaster University as a lecturer in Environmental Remote Sensing. Metabarcoding now allows ecologists to gather huge amounts of high-quality data on which species occur where; and perhaps most importantly, it is cheaper than traditional identification studying specimens under a microscope.

“Metabarcoding essentially reads millions of copies of DNA sequences that we have extracted from our samples, and those sequence reads can then be matched to a reference library that indicates what was likely to have been present in our sample.

It’s fast, the accuracy is continually improving and the cost is continually declining. Dedicated labs can process thousands of samples per year, and metabarcoding has been shown to perform as well as, if not better than, traditional survey techniques.”

But even with thousands of samples, biological surveys can only cover a small fraction of the landscape. To understand the bigger picture, Alex uses remote sensing - images from satellites, planes and drones - in his work. “Some ecological changes are easier to observe. For example we can now monitor changes in tree cover, and hence the rate of deforestation globally from satellites in near-real time”.

“However, there are very few specific plants or animals we can directly observe from aerial imagery. In fact, there are many species researchers struggle to identify, and as a result some groups are ignored entirely. Metabarcoding, and other approaches to sequencing DNA, represent a massive step forward to overcoming that issue.”

Alex is excited by the potential of combining the detailed data provided by metabarcoding with the larger scale information gained through remote sensing.

In 2017 Alex and his colleagues argued the benefit of this approach, showing that with the arrival of high-throughput methods like metabarcoding, we now have the tools we need to connect on-ground biological information to remotely sense environmental data at large scales.

He has spent the past 3 years on a major Canadian project testing the use of metabarcoding, which, along with environmental information from remote sensing will be used to assess freshwater habitat condition in Atlantic Canada, an area the size of England, as well as the wilderness of Wood-Buffalo National Park in northern Alberta [pictured].

“I think Canada is still leading the field in metabarcoding, but given the enthusiasm those early positive results have generated, and the number of European projects now looking at metabarcoding, I don’t expect that to last.” said Alex. ”Understanding how landscapes can be made more sustainable for both people and wildlife requires expertise in many areas and I think the Lancaster Environment Centre has the breadth to make such ambitious projects possible.”

“There is so much potential to go further, helping us manage pollinators, detect invasive species, or disease vectors like mosquitoes, as well as a better understanding of how ecosystems are structured. The work in Canada spanned eight government departments, covering issues from environmental health to agricultural productivity and public health, and I have high hopes for developing new projects with colleagues at Lancaster.”